Concrete Cracks on Mezzanine 4

[SteelPE]

I am being asked to visit a project that was completed about 5 years ago. I designed the project. The owner called me today after getting my name from the architect. They said that late last week they heard a loud “bang” on the second floor “followed by what sounded like ice cracking. They are now concerned about the second floor of the building.

The existing structure is a single-story building with an interior mezzanine. A portion of the mezzanine is used for storage and the remaining portion of the mezzanine is used for general office. Construction is a 3” concrete slab on 9/16” form deck with open-web steel joists designed to be spaced at 2’-6” o.c. (this area actually ended up with open web joists at 2’-4” o.c.). The joists are supported by a system of WF steel beam and steel columns.

The owner provided me with a sketch of where they say the mezzanine has cracked (it’s terrible). From this sketch the cracks appear to be happening at the 1/3 points of the joists. Also according to the sketch the cracks are not in a distinct pattern (parallel to joists, perpendicular to joists and diagonal to joists).

It seems kind of odd that this would happen 4-5 years later. Does anyone have any insight to what may be happening before I make my trip to the site?

 

[r13]

You should talk to the architect, and charge him, as he was in charge of this project, I guess.

 

[hokie66]

retired13,

Wishful thinking, much? Architects don't understand steel and concrete, they just want it to be perfect, and the engineer to make it so.

 

[SlideRuleEra]

SteelPE - Seems you made good use of your time on site.
Concerning the letter requested by the Owner, you mentioned the project, which you designed, was completed about 5 years ago. Check the statute of repose for your state. If the limit has not been reached, you may want to talk to a lawyer, cooperate (free) with the Owner, or both.

http://www.SlideRuleEra.net

 

[r13]

hokie66,

I don't know what impression you got from my comment. I though originally OP was delegated/subcontracted by the architect of structural works for the project, not directly hired by the owner. And now he was contacted, again, through the architect indirectly. I think he shall keep the architect in the loop, as he's the actual link between the owner and OP. If the owner refuses/unwilling to pay, then he shall seeks compensation from the architect. I know, it's not likely, but that is what I would do.

 

[bones206]

Say the slab is 3" NWC f'c = 4000 psi. Modulus of rupture can be estimated as f[sub]r[/sub] =5(4000)[sup]1/2[/sup] = 316 psi. Section modulus = (12")(3")[sup]2[/sup]/6 = 18 in[sup]3[/sup]/ft. Estimated cracking moment M[sub]cr[/sub] = f[sub]r[/sub]S = 5.688 in-kip/ft = 0.474 ft-kip/ft.

I did a quick RISA model with the joists omitted and ran a self-weight only case and a 20 psf live load case. The results show that it's within the realm of possibility that the slab could span between girders elastically up to a certain amount of load.

Attached the RISA file if anyone else wants to mess around with it. You could also model virtual joists based on the load-dependent moment of inertia of each bar joist and see what kind of load distribution you get.

 

[dik]

Thanks Bones206: I'm corrected... never would have guessed it.


Dik

 

[SteelPE]

bones206

I understand what you are saying. Add in the fact that the system is probably stronger than what you even show. In order for the span to span between girders deflection must happen. During the initial concrete pour the deck can't span between the girders to the slab ends up bearing on the joists. As the slab is loaded the system will deflect increasing the load in the floor joists. At this point, the slab is trying to force continuity between the joists and the girders..... until cracking occurs and then the joist end up taking the full load.

I think this is a possible scenario in a world where shrinkage cracks do not occur. Some of the observed cracks were old so I don't necessarily think this is a possible scenario.

This could go a number of different ways. I feel I have done the best I can while minimizing damages on my end (in terms of lost time). I am glad I was able to visit the site and confirm the existing loading conditions. I don't know how much additional time I want to sink into this going forward (especially since the owners mood changed once I mentioned compensation). I suppose it's fun to try an figure these things out, but fun doesn't pay the mortgage.

 

[bones206]

I guess I was being too conservative with the slab taking it's own self-weight in the analysis. That stress would already be locked into the deck and joists. So the live load only stress in the slab would be much less than what I show above. So yes the structure would be starting in an initial deformed shape, but the slab could still span across once the concrete cures and resist any superimposed loads in that fashion.

Shrinkage cracks are almost a certainty since there are a few re-entrant corners (and I'm assuming no control joints). Maybe even the weld washers cause enough restraint to induce shrinkage cracking. But this whole thought experiment is based on the premise that there was some type of sudden cracking event that the workers described. So if we accept that actually happened, then I think shrinkage cracking would be secondary and maybe explain some of the older hairline cracks.

Even if this unintended spanning theory turned out to be what happened, it doesn't mean the structure has a design problem or performed poorly. I wouldn't consider a little load redistribution to be a structural failure. It is pretty fun sometimes to noodle on theories on concepts online here, especially when it's not my project. I tend to learn a lot that way and who knows, maybe it will pay my bills someday.

 

[r13]

bones206,

Can you check unbalanced load condition, with a limited area loaded with the maximum file weight (50 psf, or 90 psf), and the adjacent area without any load in addition to slab self weight. Can be interesting to see the effect due to positive span moment.

 

[r13]

While the theory of shrinkage crack is plausible, I still think there are some other possibilities that causing the randomly orientate cracks, especially given considerations to the method of construction - wet concrete on form deck that less likely to lose water than concrete on other type of formworks, or on subgrade; and the construction of mezzanine floor usually occurs after the building envelop is up, thus the temperature effect is less a problem. So I suggest to evaluate unbalanced load conditions (using actual load conditions - heavy storage load locally, no live load nearby) to check the effect of potential moment reversal. Not necessary happening, but worth to look into.

 

[bones206]

retired13 - is this what you were looking for? I also ran a case with superimposed live load only if we were to assume all the dead load stress was taken by the decking and framing.

 

[r13]

bones206,

Thank you very much for going extra mile to verify my concern, although the result wasn't as anticipated. So the localized storage has very limited effect on the slab.

 

[KootK]

@bones206: thanks for all of the exploration that you've done with respect to our shared theory. It's been a treat following along. In addition to considering the uncracked flexural strength of the concrete, I'd also imagined that, up to a certain level, it would behave compositely with the metal deck. Some interesting features of that mechanism:

1) It could potentially be in play even after shrinkage cracking.

2) It could potentially increase slab spanning capability beyond what you've demonstrated.

3) In my head I see the loss of composite action via an "un-peeling" of the deck being a good candidate for the ice cracking noise observed by the tenants. You know, if they're not full of crap to begin with.

@SteelPE: I fully understand that you've moved on from this particular theory. I'm just having fun with it now.

 

[r13]

Point 3 of KootK's comments is quite interesting, and could explain the "noise".

 

[bones206]

I have no idea what sound a slab unpeeling from a deck would sound like, but I can believe that a nominal amount of composite action can develop then fail after a certain point. The weld washers could act as little baby shear connectors that could perhaps locally crush/crack the slab at a certain load level. Or maybe the washers would fracture off the weld first, releasing the composite action. For such a simple structure, the serviceability aspects are pretty interesting and complex.

 

[r13]

what sound a slab unpeeling from a deck would sound like

A high pitch popping sound, when an area of steel deck separated from the slab in a sudden manner, and breaking of the concrete, I guess.

 

[Tomfh]

Delaminating decking seems a reasonable suggestion for an ice cracking noise.

 

[SteelPE]

I appreciate the theories, but how do you go about proving something like that? Things may work out on paper, but to experience sudden cracking of the slab due to plan bending or to experience the concrete unbonding from the decking years later is tough to prove. I suppose seeing no evidence of damage (flat bottom chords of joists, no drywall cracking, no stuck doors etc) other than cracks in the mezzanine would help reinforce the theory. I could ask a few site supers if they have ever heard similar noises in the past. My guess is with how loud projects are the answer would be no.

What I get a little nervous about is throwing some of these ideas out there only to look like an idiot when some other engineer comes in to prove me wrong. For some reason I always picture myself having to explain my theories/designs in a deposition (this happens with most of what I do). I am not sure how far the theories would go before being dismissed.

FYI, it's been almost a week and I haven't heard from the owner. I am guessing they don't want an official report because they don't want to compensate me for my time.

 

[bones206]

First off, I feel like you've already done your due diligence with the site inspection, and we're all just armchair quarterbacks here playing with theories. I think it's healthy to put multiple hypothesis on the table then test each one to try to prove/disprove. As long as you're not irrationally falling in love with one theory and ignoring other possibilities, you're not going to look like an idiot. But I agree we all dread the idea of having to explain ourselves in the proverbial hot seat. And we do live in a litigious world so it's a rational thing to worry about. Just make sure you stick to the facts and evidence and don't try to downplay or conceal anything.

From a business perspective though, I think it would be in your best interest to keep the owner happy. I think if I was in your position, I'd send them a very short report with your findings that hopefully leaves them with some peace of mind and satisfaction that their concerns have been evaluated and addressed in a professional manner. Then I would bill the architect like retired13 suggested. Let the architect be the bad guy and bill the owner. I just don't think it's a good look to not provide a report after being called in to do a site inspection.

 

[bob33]

Has "bolt banging" been ruled out? See Rabbit 12 - 25 Feb 21:43.